mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Jason-JH Lin (林睿祥)" <Jason-JH.Lin@mediatek.com>
To: "CK Hu (胡俊光)" <ck.hu@mediatek.com>,
	"Shawn Sung (宋孝謙)" <Shawn.Sung@mediatek.com>,
	"chunkuang.hu@kernel.org" <chunkuang.hu@kernel.org>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-mediatek@lists.infradead.org"
	<linux-mediatek@lists.infradead.org>,
	"Bibby Hsieh (謝濟遠)" <Bibby.Hsieh@mediatek.com>,
	"jason-ch.chen@mediatek.corp-partner.google.com"
	<jason-ch.chen@mediatek.corp-partner.google.com>,
	"Nancy Lin (林欣螢)" <Nancy.Lin@mediatek.com>,
	"daniel@ffwll.ch" <daniel@ffwll.ch>,
	"p.zabel@pengutronix.de" <p.zabel@pengutronix.de>,
	"dri-devel@lists.freedesktop.org"
	<dri-devel@lists.freedesktop.org>,
	"airlied@gmail.com" <airlied@gmail.com>,
	"sean@poorly.run" <sean@poorly.run>,
	"matthias.bgg@gmail.com" <matthias.bgg@gmail.com>,
	"linux-arm-kernel@lists.infradead.org"
	<linux-arm-kernel@lists.infradead.org>,
	"angelogioacchino.delregno@collabora.com"
	<angelogioacchino.delregno@collabora.com>
Subject: Re: [PATCH v9 16/21] drm/mediatek: Support "Pre-multiplied" blending in OVL
Date: Tue, 18 Jun 2024 17:00:25 +0000	[thread overview]
Message-ID: <ca5f0f155d9dec0b6c853128d9fd753042a5d25e.camel@mediatek.com> (raw)
In-Reply-To: <a966988a520adc5aee5ee70657efd778d011ded0.camel@mediatek.com>

Hi CK,

On Fri, 2024-06-14 at 05:42 +0000, CK Hu (胡俊光) wrote:
> On Fri, 2024-06-14 at 10:46 +0800, Shawn Sung wrote:
> > From: Hsiao Chien Sung <shawn.sung@mediatek.com>
> > 
> > Support "Pre-multiplied" alpha blending mode on in OVL.
> > Before this patch, only the "coverage" mode is supported.
> > 
> > Signed-off-by: Hsiao Chien Sung <shawn.sung@mediatek.com>
> > ---
> >  drivers/gpu/drm/mediatek/mtk_disp_ovl.c | 42 ++++++++++++++++++++-
> > ----
> >  1 file changed, 34 insertions(+), 8 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/mediatek/mtk_disp_ovl.c
> > b/drivers/gpu/drm/mediatek/mtk_disp_ovl.c
> > index 6567806cf4e2..47d0b039a616 100644
> > --- a/drivers/gpu/drm/mediatek/mtk_disp_ovl.c
> > +++ b/drivers/gpu/drm/mediatek/mtk_disp_ovl.c
> > @@ -52,8 +52,12 @@
> >  #define GMC_THRESHOLD_HIGH	((1 << GMC_THRESHOLD_BITS) / 4)
> >  #define GMC_THRESHOLD_LOW	((1 << GMC_THRESHOLD_BITS) / 8)
> >  
> > +#define OVL_CON_CLRFMT_MAN	BIT(23)
> >  #define OVL_CON_BYTE_SWAP	BIT(24)
> > -#define OVL_CON_MTX_YUV_TO_RGB	(6 << 16)
> > +
> > +/* OVL_CON_RGB_SWAP works only if OVL_CON_CLRFMT_MAN is enabled */
> > +#define OVL_CON_RGB_SWAP	BIT(25)
> > +
> >  #define OVL_CON_CLRFMT_RGB	(1 << 12)
> >  #define OVL_CON_CLRFMT_ARGB8888	(2 << 12)
> >  #define OVL_CON_CLRFMT_RGBA8888	(3 << 12)
> > @@ -61,6 +65,11 @@
> >  #define OVL_CON_CLRFMT_BGRA8888	(OVL_CON_CLRFMT_ARGB8888 |
> > OVL_CON_BYTE_SWAP)
> >  #define OVL_CON_CLRFMT_UYVY	(4 << 12)
> >  #define OVL_CON_CLRFMT_YUYV	(5 << 12)
> > +#define OVL_CON_MTX_YUV_TO_RGB	(6 << 16)
> > +#define OVL_CON_CLRFMT_PARGB8888 ((3 << 12) | OVL_CON_CLRFMT_MAN)
> 
> #define OVL_CON_CLRFMT_PARGB8888 (OVL_CON_CLRFMT_RGBA8888 |
> OVL_CON_CLRFMT_MAN)
> 
> But I'm confused with the naming.
> 

We found that original naming is not correct, so we fix it with the
patch:

https://patchwork.kernel.org/project/linux-mediatek/patch/20240616-mediatek-drm-next-v1-11-7e8f9cf785d8@mediatek.com/

PARGB8888 is not ARGB8888 + MAN.

If DISP_REG_OVL_CON open the MAN bit, some original color format
settings will be changed, for example:

When MAN = 0:
RGB565 = 0000 in bit[15:12]
RGB888 = 0001 in bit[15:12]
ARGB8888 = 0010 in bit[15:12]
RGBA8888 = 0011 in bit[15:12]
UYVY = 0100 in bit[15:12]
YUYV = 0101 in bit[15:12]

When MAN = 1:
RGB888 = 0000 in bit[15:12]
RGB565 = 0001 in bit[15:12]
BGRA
8888 = 0010 in bit[15:12]
PARGB8888 = 0011 in bit[15:12]BGRA4444 = 0100
in bit[15:12]
BGRA5551 = 0101 in bit[15:12]
YUYV = 1000 in bit[15:12]
UYVY
= 1001 in bit[15:12]
YVYU = 1010 in bit[15:12]
VYUY = 1011 in bit[15:12]

> 
> > +#define OVL_CON_CLRFMT_PABGR8888 (OVL_CON_CLRFMT_PARGB8888 |
> > OVL_CON_RGB_SWAP)
> 
> #define OVL_CON_CLRFMT_PABGR8888 (OVL_CON_CLRFMT_ABGR8888 |
> OVL_CON_CLRFMT_MAN)
> 
> > +#define OVL_CON_CLRFMT_PBGRA8888 (OVL_CON_CLRFMT_PARGB8888 |
> > OVL_CON_BYTE_SWAP)
> 
> #define OVL_CON_CLRFMT_PBGRA8888 (OVL_CON_CLRFMT_BGRA8888 |
> OVL_CON_CLRFMT_MAN)
> 
> > +#define OVL_CON_CLRFMT_PRGBA8888 (OVL_CON_CLRFMT_PABGR8888 |
> > OVL_CON_BYTE_SWAP)
> >  #define OVL_CON_CLRFMT_RGB565(ovl)	((ovl)->data->fmt_rgb565_is_0 ?
> > \
> >  					0 : OVL_CON_CLRFMT_RGB)
> >  #define OVL_CON_CLRFMT_RGB888(ovl)	((ovl)->data->fmt_rgb565_is_0 ?
> > \
> > @@ -74,6 +83,8 @@
> >  #define	OVL_CON_VIRT_FLIP	BIT(9)
> >  #define	OVL_CON_HORZ_FLIP	BIT(10)
> >  
> > +#define OVL_COLOR_ALPHA		GENMASK(31, 24)
> > +
> >  static inline bool is_10bit_rgb(u32 fmt)
> >  {
> >  	switch (fmt) {
> > @@ -298,7 +309,13 @@ void mtk_ovl_config(struct device *dev,
> > unsigned int w,
> >  	if (w != 0 && h != 0)
> >  		mtk_ddp_write_relaxed(cmdq_pkt, h << 16 | w, &ovl-
> > >cmdq_reg, ovl->regs,
> >  				      DISP_REG_OVL_ROI_SIZE);
> > -	mtk_ddp_write_relaxed(cmdq_pkt, 0x0, &ovl->cmdq_reg, ovl->regs, 
> > DISP_REG_OVL_ROI_BGCLR);
> > +
> > +	/*
> > +	 * The background color must be opaque black (ARGB),
> > +	 * otherwise the alpha blending will have no effect
> > +	 */
> > +	mtk_ddp_write_relaxed(cmdq_pkt, OVL_COLOR_ALPHA, &ovl-
> > >cmdq_reg,
> > +			      ovl->regs, DISP_REG_OVL_ROI_BGCLR);
> 
> The coverage mode formula is:
> dst.RGB = src.RGB * src.A + dst.RGB * (1 - src.A)
> 
> The pre-multiplied mode formula is:
> dst.RGB = src.RGB + dst.RGB * (1 - src.A)
> 
> Both formula has no destination alpha (I think background color is
> last destination),
> why coverage mode work fine but pre-multiplied mode has something
> wrong?
> 

alpha of coverage mode formula is:
dst.a = dst.a x (0xff - src.a x SCA/0xff)/0xff + src.a x SCA/0xff;

alpha of pre-multiplied mode is:
dst.a = round( {{dst.a x [0xff - round((src.a x SCA)/0xff)]} / 0xff} +
round((src.a x SCA)/0xff) );

* SCA is constant alpha

If SCA = 0xff, then the formula will be:
dst.a = dst.a x (0xff - src.a) + src.a

If dst.a is 0, then dst.a will be src.a
If dst.a is 0xff, then dst.a will be 0xff


Regards,
Jason-JH Lin

> Regards,
> CK
> 
> >  
> >  	mtk_ddp_write(cmdq_pkt, 0x1, &ovl->cmdq_reg, ovl->regs,
> > DISP_REG_OVL_RST);
> >  	mtk_ddp_write(cmdq_pkt, 0x0, &ovl->cmdq_reg, ovl->regs,
> > DISP_REG_OVL_RST);
> > @@ -374,7 +391,8 @@ void mtk_ovl_layer_off(struct device *dev,
> > unsigned int idx,
> >  		      DISP_REG_OVL_RDMA_CTRL(idx));
> >  }
> >  
> > -static unsigned int ovl_fmt_convert(struct mtk_disp_ovl *ovl,
> > unsigned int fmt)
> > +static unsigned int ovl_fmt_convert(struct mtk_disp_ovl *ovl,
> > unsigned int fmt,
> > +				    unsigned int blend_mode)
> >  {
> >  	/* The return value in switch "MEM_MODE_INPUT_FORMAT_XXX"
> >  	 * is defined in mediatek HW data sheet.
> > @@ -395,22 +413,30 @@ static unsigned int ovl_fmt_convert(struct
> > mtk_disp_ovl *ovl, unsigned int fmt)
> >  	case DRM_FORMAT_RGBA8888:
> >  	case DRM_FORMAT_RGBX1010102:
> >  	case DRM_FORMAT_RGBA1010102:
> > -		return OVL_CON_CLRFMT_RGBA8888;
> > +		return blend_mode == DRM_MODE_BLEND_COVERAGE ?
> > +		       OVL_CON_CLRFMT_RGBA8888 :
> > +		       OVL_CON_CLRFMT_PRGBA8888;
> >  	case DRM_FORMAT_BGRX8888:
> >  	case DRM_FORMAT_BGRA8888:
> >  	case DRM_FORMAT_BGRX1010102:
> >  	case DRM_FORMAT_BGRA1010102:
> > -		return OVL_CON_CLRFMT_BGRA8888;
> > +		return blend_mode == DRM_MODE_BLEND_COVERAGE ?
> > +		       OVL_CON_CLRFMT_BGRA8888 :
> > +		       OVL_CON_CLRFMT_PBGRA8888;
> >  	case DRM_FORMAT_XRGB8888:
> >  	case DRM_FORMAT_ARGB8888:
> >  	case DRM_FORMAT_XRGB2101010:
> >  	case DRM_FORMAT_ARGB2101010:
> > -		return OVL_CON_CLRFMT_ARGB8888;
> > +		return blend_mode == DRM_MODE_BLEND_COVERAGE ?
> > +		       OVL_CON_CLRFMT_ARGB8888 :
> > +		       OVL_CON_CLRFMT_PARGB8888;
> >  	case DRM_FORMAT_XBGR8888:
> >  	case DRM_FORMAT_ABGR8888:
> >  	case DRM_FORMAT_XBGR2101010:
> >  	case DRM_FORMAT_ABGR2101010:
> > -		return OVL_CON_CLRFMT_ABGR8888;
> > +		return blend_mode == DRM_MODE_BLEND_COVERAGE ?
> > +		       OVL_CON_CLRFMT_ABGR8888 :
> > +		       OVL_CON_CLRFMT_PABGR8888;
> >  	case DRM_FORMAT_UYVY:
> >  		return OVL_CON_CLRFMT_UYVY | OVL_CON_MTX_YUV_TO_RGB;
> >  	case DRM_FORMAT_YUYV:
> > @@ -450,7 +476,7 @@ void mtk_ovl_layer_config(struct device *dev,
> > unsigned int idx,
> >  		return;
> >  	}
> >  
> > -	con = ovl_fmt_convert(ovl, fmt);
> > +	con = ovl_fmt_convert(ovl, fmt, blend_mode);
> >  	if (state->base.fb) {
> >  		con |= OVL_CON_AEN;
> >  		con |= state->base.alpha & OVL_CON_ALPHA;

  reply	other threads:[~2024-06-18 17:00 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-06-14  2:45 [PATCH v9 00/21] Support IGT in display driver Shawn Sung
2024-06-14  2:46 ` [PATCH v9 01/21] soc: mediatek: Disable 9-bit alpha in ETHDR Shawn Sung
2024-06-14  2:46 ` [PATCH v9 02/21] drm/mediatek: Add OVL compatible name for MT8195 Shawn Sung
2024-06-14  2:46 ` [PATCH v9 03/21] drm/mediatek: Add missing plane settings when async update Shawn Sung
2024-06-14  2:46 ` [PATCH v9 04/21] drm/mediatek: Add DRM_MODE_ROTATE_0 to rotation property Shawn Sung
2024-06-14  2:46 ` [PATCH v9 05/21] drm/mediatek: Set DRM mode configs accordingly Shawn Sung
2024-06-14  2:46 ` [PATCH v9 06/21] drm/mediatek: Turn off the layers with zero width or height Shawn Sung
2024-06-14  2:46 ` [PATCH v9 07/21] drm/mediatek: Support more 10bit formats in OVL Shawn Sung
2024-06-14  2:46 ` [PATCH v9 08/21] drm/mediatek: Support RGBA8888 and RGBX8888 in OVL on MT8195 Shawn Sung
2024-06-14  2:46 ` [PATCH v9 09/21] drm/mediatek: Fix XRGB setting error in OVL Shawn Sung
2024-06-14  6:06   ` CK Hu (胡俊光)
2024-06-14  2:46 ` [PATCH v9 10/21] drm/mediatek: Fix XRGB setting error in Mixer Shawn Sung
2024-06-14  3:58   ` CK Hu (胡俊光)
2024-06-14  2:46 ` [PATCH v9 11/21] drm/mediatek: Add new color format MACROs in OVL Shawn Sung
2024-06-14  4:04   ` CK Hu (胡俊光)
2024-06-14  2:46 ` [PATCH v9 12/21] drm/mediatek: Support DRM plane alpha " Shawn Sung
2024-06-14  5:06   ` CK Hu (胡俊光)
2024-06-14  2:46 ` [PATCH v9 13/21] drm/mediatek: Support DRM plane alpha in Mixer Shawn Sung
2024-06-14  5:08   ` CK Hu (胡俊光)
2024-06-14  2:46 ` [PATCH v9 14/21] drm/mediatek: Support "None" blending in OVL Shawn Sung
2024-06-14  5:11   ` CK Hu (胡俊光)
2024-06-14  2:46 ` [PATCH v9 15/21] drm/mediatek: Support "None" blending in Mixer Shawn Sung
2024-06-14  5:13   ` CK Hu (胡俊光)
2024-06-14  2:46 ` [PATCH v9 16/21] drm/mediatek: Support "Pre-multiplied" blending in OVL Shawn Sung
2024-06-14  5:42   ` CK Hu (胡俊光)
2024-06-18 17:00     ` Jason-JH Lin (林睿祥) [this message]
2024-06-19  3:17       ` CK Hu (胡俊光)
2024-06-19 16:00         ` Jason-JH Lin (林睿祥)
2024-06-14  5:49   ` CK Hu (胡俊光)
2024-06-14  2:46 ` [PATCH v9 17/21] drm/mediatek: Support "Pre-multiplied" blending in Mixer Shawn Sung
2024-06-14  5:55   ` CK Hu (胡俊光)
2024-06-14  2:46 ` [PATCH v9 18/21] drm/mediatek: Support alpha blending in display driver Shawn Sung
2024-06-14  2:46 ` [PATCH v9 19/21] drm/mediatek: Support CRC " Shawn Sung
2024-06-14  2:46 ` [PATCH v9 20/21] drm/mediatek: Support CRC in OVL Shawn Sung
2024-06-14  2:46 ` [PATCH v9 21/21] drm/mediatek: Support CRC in OVL adaptor Shawn Sung

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=ca5f0f155d9dec0b6c853128d9fd753042a5d25e.camel@mediatek.com \
    --to=jason-jh.lin@mediatek.com \
    --cc=Bibby.Hsieh@mediatek.com \
    --cc=Nancy.Lin@mediatek.com \
    --cc=Shawn.Sung@mediatek.com \
    --cc=airlied@gmail.com \
    --cc=angelogioacchino.delregno@collabora.com \
    --cc=chunkuang.hu@kernel.org \
    --cc=ck.hu@mediatek.com \
    --cc=daniel@ffwll.ch \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=jason-ch.chen@mediatek.corp-partner.google.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mediatek@lists.infradead.org \
    --cc=matthias.bgg@gmail.com \
    --cc=p.zabel@pengutronix.de \
    --cc=sean@poorly.run \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®